SuperNode( aka SuperPod/HyperNode)
kubernetes-sigs/lws#620
Scaling your LLM inference workloads: multi-node deployment with TensorRT-LLM and Triton on Amazon EKS
https://aws.amazon.com/cn/blogs/hpc/scaling-your-llm-inference-workloads-multi-node-deployment-with-tensorrt-llm-and-triton-on-amazon-eks/
- huawei CloudMatrix 384
- NVIDIA DGX nvl72 / GB200
- ScaleX640
- Alibaba's Panjiu AI Infra 2.0 128 cards
as a topo key?
